US3132156A

US3132156A - Selective non-catalytic, vapor phase oxidation of saturated aliphatic hydrocarbons to olefin oxides

Info

Publication number: US3132156A
Application number: US66469A
Authority: US
Inventors: Russel C Lemon; Philip C Johnson; Jozsef M Berty
Original assignee: Union Carbide Corp
Current assignee: Union Carbide Corp
Priority date: 1960-11-01
Filing date: 1960-11-01
Publication date: 1964-05-05
Anticipated expiration: 1981-05-05
Also published as: GB970878A

Description

May 5, 1964 R. c. LEMON ETAL SELECTIVE NON-CATALYTIC, VAPOR PHASE OXIDATIO OF SATURATED ALIPHATIC HYDROCARBONS T0 OLEFIN OXIDES 6 Sheets-Sheet 1 Filed Nov. 1, 1960 Hydrocarbon Feed g\ STIRRED POT Hydrocarbon Feed POT WITH VENTURI MIXING NOZZLE Product Gas 26 DOUBLE CONE REACTOR O b MO) r-ll m 0 n 0 0 0 w w r M m/mm d a F n o b r a C o r d y H N Sw R o 0 N T THMR m M V J mC M HUO " vis/wry IDATION ARBONS 5, 1964 'R. c. LEMON SELECTIVE NON ETAL VAPOR PHASE OX OF SATURATED ALIPHATIC HYDROC CATALYTIC,

TO OLEFIN OXIDES Filed Nov. 1, 1960 6 Sheets-Sheet 3 v 500 4 CENTIGRADE w m w m m 81328 281385: 8 3 Q: @233 3:6 UZSEQ E 5565 a0 PRESSURE PSLG.

M 82 328 mzomfiuw i u 8 2: 853 z. :56 596% 2 555::

I00 I20 I 60 0 PRESSURE PSLG.

o mzu4 momm INVENTORS PHILIP C.JOHNSON RUSSELC. LEMON JOZS EF M. BERTY av a ATTORN Y May 1954 R LEMON ETAL 3132156 C- SELECTIVE NON-CATALYTIC, VAPOR PHASE OXIDATION OF SATURATED ALIPHATIC HYDROCARBONS TO OLEFIN OXIDES Filed Nov. 1, 1960 6 Sheets-Sheet 4 YIELD IN MILLIGRAMS/LITER OF PROPYLENE OXIDE IN PRODUCT STREAM O 20 40- so ao 10o VOLUME OF PROPYLENE IN FEED By d0. A770 NEV v 3,132,156 May 5, 1964 R. c. LEMON ETAL SELECTIVE NON-CATALYTIC, VAPOR PHASE OXIDATION OF SATURATED ALIPHATIC HYDROCARBONS TO OLEF IN OXIDES Filed Nov. 1. 1960 v s Sheets-Sheet e INVENTORS PHILIP C. JOHNSON RUSSEL C. LEMON JOZSEF M. BERTY BY W ATTOR Y United States Patent WSE'LECTIVE. NoN-oArALYrrc, vAron rrrasr:

OXIDATEON 9F SATURATED ALHPHATIC HY- DROCARBONS TO GLEFIN GXRDES Russel C. Lemon, Scott Depot, Philip C. Johnson, St.

Albans, and .lozsef M. Berty, Charleston, W. Va.,

assignors to Union Carbide Corporation, a corporation of New York Filed Nov. 1, 1960, Ser. No. 66,469 12 Claims. (Cl. 260-348) This invention relates to the non-catalytic, vapor phase oxidation of saturated, aliphatic hydrocarbons containing from 2 through 4 carbon atoms and has, for one of its principal objects, the provision of an improved process for the production of propylene oxide from propane.

Alkylene oxides (vicinal epoxy alkanes), and particularly propylene oxide, are very valuable and widely used chemicals. They. have been polymerized with a wide variety of monomers to yield polymers which have been useful in coating compositions and in the manufacture of molded articles. Alkylene oxides have also been reacted with alcohols to yield monoalkyl ethers which have atced with propylene to form propylene chlorohydrin. 1

The propylene chlorohydrin is then dehydrohalogenated to yield propylene oxide. Another method to obtain propylene oxide is by the liquid phase oxidation of propylene with organic peracids.

The reactions, involved in the. aforementioned methods are, specific and although substantially only the desired product is produced, there are serious disadvantages associated therewith. The chlorohydrin process suf fers from the disadvantage of forming by-products which increase the cost of operation and necessitate elaborate,

expensive and time-consuming separation techniques.

1 Additionally, the 'raw materials, chlorine and propylene are relatively expensive and the corrosive nature of chlor'ine requires special and expensive equ1pment. The

oxidation of propylene with peracids is a potentially dangerous operation and expensive equipment is'needed to guard against the hazards of the peracids. Another disadvantage of this method is the high cost of peracids.

Another method which has received considerable attention in the literature is the direct oxidation of hydrocarbons with an oxygen-containing gas. This method suifers from one serious disadvantage in that it is not specific for the production of alkylene oxides but produces a variety of other compounds including acids, esters, ethers and oxides of carbon including carbon monoxide and carbon dioxide. The reaction does, however, possess two attributes which recommend it highly for commercial utilization, i.e., inexpensiveness of starting materials and simplicity of operation. It is primarily for these reasons that much of the attention in recent years has been directed to improvements in methods for the production of alkylene oxides from the air oxidation of hydrocarbons even though the producer must necessarily contend with the concurrent production of a variety of undesired products.

With the advent of methods for the production of alkylene oxides by the direct air oxidation of hydrocarbons, particular emphasis has been placed upon the maximization of oxide yield and the minimization of yields of other products of the reaction.

3,l32,l5 6 Patented May 5, 1964 ice In order to facilitate further understanding of the problems involved in the air oxidation of hydrocarbons as well as the improved method of this invention, references will be had hereinafter to propylene oxide made from propane. It is to be understood however, that the principles set forth are generally applicable to saturated aliphatic hydrocarbons containing from 2 through 4 carbon atoms, as pointed out above.

' The difficulty in arriving at a method which is highly selective for the production ofpropylene oxide can be illustrated by an oversimplification of the reactions which occur when propane is oxidized in the presence of oxygen or an oxygen-containing gas. For the purpose of this explanation it can be assumed that propane undergoes the following competing reactions:

carbon oxides, water, aldehydes, etc.

One reaction involves the formation of propylene oxide while another reaction involves the formation of intermediates other than propylene oxide and simultaneously therewith still another reaction involves the degradation of propylene oxide previously formed. Since the oxidation'is highly exothermic and self-accelerating, the direction of the oxidation to a partially oxidized product such as propylene oxide presents many problems.

The prior art methods which attempted to produce propylene oxide by theoxidation of propane were only partially successful. The majority of the prior art methods used conventional vertical columns and differed from each other by variations in lengths and diameter of the column, temperature, pressure, etc. However, all of these methods suffered one common disadvantage-the temperature and the concentration of the reactants varied intermediates throughout the length of the column.

The temperature variations are easily explained since the oxidation reactions are exothermic and the heat evolved differs with each reaction which is taking place.

These prior art methods necessitated the use of elaborate and expensive cooling apparatus which, unfortunately, could not provide a uniform temperature. Thus, at various increments along the tube, conditions existed which favored the direction of the oxidation to products other than propylene oxide.

Additionally, the concentration of reactants varied throughout the length of the tube due to the fact that oxi dation takes place at each ways and in various rates.

Thus, it is the primary purpose of this invention to point in the tube in various provide a method for the production of good yields of an alkylene oxide from the oxidation of a saturated aliphatic hydrocarbon containing from 2 through 4 carbon atoms with oxygen or an oxygen-containing gas.

It is another object of this invention to provide a methed for the direction of the oxidation of propane to propylene oxide which is efhcient, simple, and provides good yields of propylene oxide and is cheaper to install, maintain and operate.

It is still another object of this invention to provide a method for the oxidation of propane which maximizes the yields of propylene oxide and minimizes the yields of unwanted by-products.

It has'now been discovered that the above-mentioned objects can finally be attained by oxidizing saturated,

such a manner that a critical balance is maintained between temperature, pressure, oxygen, and contact time in a critical environment of reactants, products and temperature gradients in the reaction zone.

It can be seen from the above that there are two coacting, essential areas in this inventiona critical balance and a critical environment. It cannot be too strongly emphasized that both areas of criticality must exist and cooperate simultaneously in order to direct the oxidation of propane to propylene oxide in good yield and efficiencies. pressure, oxygen and contact time were to exist in an environment other than the critical environment, the process would not be eliective. In fact, it could very well be that little or no propylene oxide could be recovered.

The necessity of having a critical balance which coacts with a critical environment can be very readily appreciated when it is pointed out that in the method of this invention propylene oxide is produced at a temperature which is above the temperature at which propylene oxide is degraded.

The above statement is the main theme or key to'the absolute necessity of controlling and maintaining a critical balance in a critical environment. From a hasty first Thus, if the critical balance of temperature, 7

impression it might appear that the novel method of this invention is impossible since it appears to violate practically every known thermodynamic principle. However, such is not the case as will be shown when the correla- .tion between the critical balance and the critical environment are set forth.

The critical environment in this invention comprises two elements; (1) maintaining a condition of substantial homogeneity of reactants and reaction products and (2) maintaining essentially isothermal conditions throughout the entire reaction zone.

As used herein the expression maintaining a condition of substantial homogeneity of reactants and reaction products in the reaction zone is intended to mean that the concentration of reactants and reaction products is substantially constant throughout the entire reaction zone.

This condition can be accomplished in a variety of ways and is not dependent on, or restricted to, a specific reaction vessel and is accomplished by maintaining the reaction zone in 'a high degree of turbulence whereby the reactants and reaction products are intimately. mixed throughout said reaction zone so that there is substantially no concentration gradient therein. The high degree of turbulence can be readily obtained by stirring the reactants and reaction products with a high speed stirrer or by introducing the reactants under pressure so that sufficient turbulence is achieved or by the use of special nozzles designed to allow the entry of reactants at a high rate of flow or by any combination or" the above methods. The second element of the critical environment is the requirement that the reaction zone be maintained under substantially isothermal condition.

The expression essentially isothermal as used herein is intended to mean that the temperature remains essentially constant throughout the entire reaction zone. An isothermal condition can be maintained by removing heat from the reaction zone via heat transfer means or preferably by maintaining a substantially adiabatic condition in the reaction zone. One means of maintaining an adiabatic condition while the reaction zone is isothermal is by adding the feed gases andv any recycle gases to the reaction zone at a temperature lower than the reaction temperature in such a manner that the relatively cold incoming gases will intimately mix with all the reactants and reaction products and absorb the heat liberated by the exothermic oxidation reactions, so as to maintain the reaction temperature essentially constant.

As has been previously pointed out, the existence of the critical environment in the instant method is indethe reactor.

Suitable reactors in which the critical environment can be maintained include an ordinary stirred pot, a pot with a venturi mixing nozzle and a. double cone reactor. These various reactors are shown in FIGURES l-3.

FIG. 1 shows a simple autoclave 4 which is equipped with a mechanical stirrer 3. Oxygen is introduced through 'line 1 and the hydrocarbon feed through line 2. The oxygen and the feed are subjected to a high degree of turbulence so that there is no concentration gradient in the reaction zone 5. The product gases are removed via line 6 and the unreacted materials can be recycled (not shown). Adiabaticand isothermal conditions are maintained by introducing the reactants at a temperature which is lower than the reaction temperature so that the excess exothermic heat of reaction is absorbed.

FIG. 2 depicts another reaction'vessel which is merely a pot with a venturi mixing nozzle. In this, vessel, the gaseous reactants enter the line 11 and pass through nozzle 12 at rate suflicient to give the desired reaction pressure in chamber 13. The gases from nozzle 12 pass rapidly into the duct 14, causing a lower pressure region in the open space between nozzle 12 and duct 14, this open space is designated 15 in the diagram. Duct 14 is supported on the frame'lti by three small supports. This lower pressureregionaround space 15 causes gases in the chamber to pass into duct 14, thereby mixing the gases in the chamber 13 with incoming gases from inlet line ll. The ratio of'gases entering space 15 to gases existing from nozzle 12 can be adjusted by altering the size of the opening in nozzle 12. The specific illustrated reactor is designed to achieve a volume ratio of approximately 20 to l; i.e., 20 volumes of gas in the chamber mix, with l'volume of entering gas.

, Some of the gases entering open space 15 also pass up the column -17 in counterfiow to incoming gases from nozzle 12, and pass out of the reaction region by line 13 to be recovered by suitable means.

By the 'procedurejdescri'bed it is possible to obtain substantial homogeneity of reactants'and reaction products under essentially adiabatic conditions.

As a typical example, if propane is to be oxidized with 12 volume percent oxygen at 500 C. the temperature rise under adiabatic conditions is about 340 C. Therefore, because there is complete homogeneity, and because he oxidation need take place at 500 C., the feed gas need be heated only to C. to attain essentially isothermal and adiabatic conditions.

The reactor illustrated has a volume of 16.7 cubic inches, an inside diameter of 2 inches and a length of 6 inches.

FIG. 3 depicts still another reactor in which the critical environment of this invention can be met. The figure illustrates a double cone reactor; In this reactor two 30-cones connected at their apexes outline the geometrical shape of the double-cone reactor. .The first cone is approximately one-tenth the v'olurneof thesecond one and serves as the mixing chamber, for the oxygen. The preheated feed stream of hydrocarbons is :fed tangentially at 21 into the first cone 22 and oxygen is introduced into this rotating body of gases through a side tube 23 axially located at'the base of the cone. The centrifugal force causes intimate and instantaneous mixing of the oxygen and the resulting mixture then spins into the second the throat or the'point of connection of the two cones which causes the gas mixture near the base of the second cone to return to the apex of the cone via a current running axially at the center of the chamber. [his reverse fiow of gas causes the entire mass of gas to break up into an infinite number of small eddies and both intense mixing and back-mixing are achieved in this section. V

The high velocity of the gases through the throat prevents back propagation of the reaction into the first chamber and the reaction zone is completely confined within the second chamber. Auxiliary streams of gases for moderating the reaction can be introduced through the injection tube axially directed from the base of the second cone 25 and the products are withdrawn at line Itshould become apparent that the critical environment of this invention is one which does not favor the recover-y of propylene oxide. Thus, as has been stated, the critical environment consists of substantial homogeneity of reactants and reaction products and essentially isothermal conditions. Thus, under the environment of this invention any propylene oxide formed is intimately mixed with oxygenand should degrade; FIG. 4 shows the degradation of propylene oxide in relation to the temperature. From this graph it can be seen that, at a temperature of 500 C., propylene oxide degrades at a rate of approximately 30 percent per second.

Thus, it would appear absolutely impossible to have a method for maximizing propylene oxide which requires that method to have a critical environment which tends to destroy propylene oxide once it is formed. However, such is not the case. .It is again stressedthat there are two areas of criticality in the novel method of this invention-a' critical balance of variables in a critical environment.

In order to facilitate a better understanding of the necessity for controlling and maintaining the relationship between the critical balance and the critical environment, it is necessary to understand the problems involved in the oxidation of propane to propylene oxide.

In such an oxidation two distinct concepts must be kept clear: (1) the amount of propylene oxide which is produced in the reaction zone and (2) the amount of propylene oxide which is recovered in the product stream., Thus, at temperatures of aboutSOO C. and under homogeneous conditions a comparatively large amount of propylene oxide is formed in the reaction zone. the presence of excess oxygen and the high temperature,

However, ,due to the propylene oxide is further degraded to other. products so that the amount of propylene oxide which is recovered in the product stream is comparatively low. Conversely, when the temperature is maintained at about 350 C. and

the reaction is conducted under non-homogeneous conditions, less propylene oxide is formed in the reaction zone but very little of the propylene oxide formed isdegraded to other products so that more propylene oxide is recovered in the product stream. Thus, at high temperatures, the rate of degradation of the propylene oxide formed is so great (see FIG. 4) that extremely low yields of propylene oxide in the product stream were obtained in the prior artrnethods.

In view of the above, all the efforts in the prior art Were directed to methods which did not provide an environment which favored the maximum production of propylene oxide in the reaction zone because these methods failed to prevent the degradation of propylene oxide once it was formed. -In other words, the prior art processes provided environments which favored the stability of any propylene oxide which was formed and not to environments which favored the production of the maximum amount of propylene oxide in the .reaction zone.

In view of the above, it can readily be observed that the relation between the critical environment which favors the maximum production of propylene oxide and the critical balance ofvariables, which efliectively prevents the degradation of propylene oxide formed to any substantial degree must be carefully controlled.

As has been previously pointed out, the critical balance of variables involves the careful regulation of temperature, pressure, oxygen concentration and contact time in the reaction zone.

The necessity of regulating the amount of oxygen in the feed has its basis in a twofold consideration the amount of oxygen present which serves to oxidize propane to propylene oxide and the amount of oxygen present which serves to react with the propylene oxide formed and further oxidizes it to unwanted products.

The above-mentioned twofold consideration only sets forth the necessity of regulating the amount of oxygen and not how to regulate it. -In order to arrive at how to regulate the amount of oxygen in the feed, a clear understanding of the process of the oxidation reaction is necessary. c

Assuming that all variables in a typical propane oxidation, except the amount of oxygen in the feed, are kept constant, and that the product is withdrawn at certain intervals and the unreacted propane recycled, certain definite conclusions can be reached. t 1

Under the above-mentioned set of facts, as the concentration of oxygen increases in the feed, the yield of propylene oxide, in milligrams, per liter of product gas, also increases. This correlation is not too unexpected since the more oxygen that is present, the more propylene oxide will be present in a liter of product stream. However,-as the oxygen concentration increases, the amount of propylene oxide formed per pounds of hydrocarbon consumed decreases. Thus, two distinct relationships are observed. The amount of propylene oxide obtained per .liter of product gas is directly proportional to the oxygen concentration and the amount of propylene oxide formed per 100 pounds of hydrocarbon consumed is inversely proportional to the oxygen concentration. Although these two relationships would appear to be contradictory, such is not thecase as will be seen once the results are understood.

If propane is oxidized in the presence of excess oxygen, quite a substantial amount of propylene oxide will be present in one liter of the product stream. The reason for this is quite obvious since the presence of excess oxygen will cause more propane to become oxidized. However, there is a vast difference between the amount of propylene oxide obtained per liter of product gasand the efficiency of the oxidation reaction. As has been previously stated, the efliciency, i.e., the pounds of propyleneoxide per 100 pounds of hydrocarbon consumed, is inversely proportional to the oxygen concentration. The reason why this is so is that although more propylene oxide is formed per liter of product gas, more of the propane is oxidized to products other than propylene oxide. Additionally, because of the large concentration of oxygen, a substantial amount of the propylene oxide formed in the reaction zone is further oxidized to unwanted by-products. Therefore, it can be seen that, although more propylene oxide is formed per liter of product gas as the oxygen concentration increases, it takes more propane to form this same amount of propylene oxide then it would if a lower oxygen concentration were present. t

From the above explanation it can be seen that, in directing the oxidation of propane to a particular oxidation product, a careful balance must be maintained between the amount of propylene oxide which can be re covered from one liter of a product stream and the amount of propylene oxide which is produced from 100 pounds of hydrocarbon. The factors determining just how this balance should be maintained are just as much theoretical as they are economical. Thus, it might be theoretically possible to have an oxygen concentration sufliciently low that the amountof propylene oxide recoverable per liter of product gas would in turn be low, so that the efliciency of the reaction would be high. This conclusion can be reached by consideringthe correlation of propylene oxide production in relation'to oxygen concentration. However, it would be economically undesirable tohave too low a concentration of propylene oxide per literof prodof the total feed, a method results which produces propylene oxide in good yields with a high degree of economy. It is particularly preferred to use oxygen concentrations in the range of 6 to 8 percent by volume since at these concentrations the process is maximized from both'a theoretical and economical consideration. following table will illustrate the effect of the oxygen concentration on the production of propylene oxide. In this table experiments have been run under four different sets of conditions. It is to be noted that three examples have been run for each one of the four difierent sets of conditions. The only variations in conditions be tween the examples in each set are slight variations in temperature as can be seen from the listing ofthese temperatures and also a slight variation of propane in the feed. This variation of propane is quite obvious since the oxygen'concentration in the feed is increased. Thus, at 8 percent by volume of oxygen in the feed there is two percent less of propane then when the oxygen concentra-' tion is 6 percent.

TABLE I Yield of Yield of propylene Volume propylene oxide in Example percent or oxide in lbs. per, 100

Oz in feed mg./l. of lbs. of Ca I product hydrogas carbons consumed As was pointed out in the preceding section, oxygen has a definite effect on the oxidation of propane. However, this 'efiect is proportional to the amount of time in which oxygen is in contact not only with the feed but also with the reaction products. Thus, if a definite concentration ofoxygen were present in the feed stream the effect of this oxygen on the direction of the oxidation of propane would havea positive relationship to the amount of time the reactants and the reaction products were in the reaction zone. If one-were to keep all the variables in a propane oxidation reaction constant except for the contact time several conclusions can be reached. It would become readily apparent that the concentration of propylene oxide in milligrams per liter of product gas is directly proportional to the contact time. This relationship is very logical andvery easily seen since the identical relationship exists as between oxygen concentration and the amount of propylene oxide in the product gas. However, itis at this point that the analogy between contact time and oxygen concentration ceases. It would be expected that at longer contacttimes the efficiency of the oxidation of hydrocarbons to propylene oxide would decrease. However, such is not the case. The eifect of contact time on the efficiency of the reaction is negligible. Thus, it is The preferred to carryout thereaction at relatively long contact timessince more propylene oxide will be present in the product stream at each definite interval of time and the'efficiency of the reaction will not sufier.

Translating the above-mentioned theoretical considerations into actual units of contact time, it has now been found that the objects of this invention can be achieved if i the oxidation reaction is carriedout when the contact times are from 0.07 to 1.5 seconds.

The preferred range is from 0.10 to 0.6 second.

The following table will illustrate the relationship of contact time to the yield of propylene oxide. The table depicts four sets of experiments which were conducted at various conditions within the scope of this invention. The individual experiments in each set were conducted at the same conditions except for the variation in contact times.

TABLE II Milligrams Lbs. of r of propylene Example O. Time in propylene oxide per seconds oxide per lbs. liter of of C3 hyproduct drocarbous gas consumed From the above table it can be seen that the yield of propylene oxide in milligrams per liter of product gas was directly proportional to the contact time, whereas the efliciencyremained practically constant in each set of experiments.

FIG. 5 represents a graph of the contact time versus the efficiency. It can be seen that the efiiciency, at each level of oxygen concentration remains almost constant. Another variable which is partof the critical balance is temperature. As has been previously pointed out, the process of this invention is operated at temperatures which are above the degradation temperature of propylene oxide. Thus, it is absolutely essential that the process of this invention be conducted at temperatures within the range of about 425 to 575 C. The'preferred range is from 450-550 C.

FIG. 6 represents a graph of the yield of propylene oxide per liter of product gas and FIG. 7 shows the relationship of efficiency to temperature." Both these graphs dramatically illustrate the better results obtained as the temperature increases. These graphs are based on experiments conducted in a double cone reactor at a pressure of 45 p.s.i.g., a contact'time of 0.12 second and a feed which consisted of: v

- Percent by volume Propane 62 Propylene 20 Carbon monoxide 10 Oxygen 8 150'p.s.i.g. The preferred range for the production of propylene oxide is from 30 to 75 p.s.i.g.

percent carbon monoxide by volume 8 percent oxygen by volume percent propylene by volume 62 percent propane by volume As has been constantly stressed heretofore, the method of this invention requires the maintaining and controlling of a critical balance of variables in a critical environment of the reaction zone. In view of the discussion of these areas of criticality, the above statement can be translated into actual numbers. It is absolutely critical in the process of this invention to operate at a pressure of 20 to 150 p.s.i.g. at a temperature of 425-575 C., at an oxygen con-. centration of 4-14 percent by volume at a contact time of 0.07 to 2 seconds in a reaction zone in which there is substantial homogeneity ofreact'ants and reaction products and essentially isothermal condition.

' As has been stated, the process of this invention is applicable to hydrocarbons which have from 2 to 4 carbon atoms. Although these hydrocarbons. can be oxidized alone, it is preferred to add a certain amount of the corresponding olefin to the feed since the results are improved. Thus, when propane is being oxidized to propylene oxide, the addition of some propylene enhances the results. The amount of olefin added can vary over a wide range but it is preferred to employ a mol ratio of saturated hydrocarbon to olefin of from 1:2 to 1(lzl.

The particularly preferred mol ratiois from 3 :1 to 4: 1.

- In another aspect of this invention, it is preferred to FIGURE 10 and FIGURE 11 are graphs of the yield and efiiciency of a propane oxidation to propylene oxide. These graphs are based on experiments conducted in a double cone reactor at a pressure of 45 p.s.i.g., an oxygen concentration of '12 percent by volume, a contact time of 1.2 seconds, and a temperature of 500 C.

Although one of the primary objects of this invention is to provide a method for the direction of the oxidation of propane to propylene oxide to obtain good yields and efficiencies, still another aspect of this invention deals with controlling the yield and efficiency of the oxidation reaction to acetaldehyde.

As has been previously stated there can be many oxidation products when propane is oxidized. Also, it is of the utmost importance to maximize the yield of the desired product. This in itself is not the best answer in arriving at ahighly economical process. The reason for this is that some of the unwanted by-products in a propane oxidation are much more easily removed from the product stream than others. One of the most difficult lay-products to remove is acetaldehyde. Thus, it would be a tremendous advantageto obtain a process which not only maximizes the yield and efiiciency of the reaction to propylene oxide but also decreasesthe amount of acetaldehyde produced per 100 pounds of hydrocarbons consumed.

The method of this invention can accomplish this result. It is immediately evident that such a result is a tremendous improvement over the prior art.

{It has now been found that at higher temperatures within the critical temperature range of this invention, efficiency of acetaldehyde decreases. This fact is shown in FIG. 12 where the efficiency of acetaldehyde is plotted 1 against the temperature. This graph was arrived at from j experiments conducted in a double cone reactor at a incorporate an inert gas into the feed stream. The inert gas serves to aid in controlling the rate of the oxidation in a way which is not completely understood. The gases which can be employed are carbon monoxide, carbon dioxide, nitrogen, helium, etc. The inert gas is used in the amount of about 10 percent by volume of the total feed mixture. The following table will. illustrate the results obtained by the addition of propylene and carbon monoxide to the feed in three sets" of experiments. The experiments in each set are conducted under similar conditions except for the feed stream.

TABLE III Feed, percent by volume Propylene Propylene oxide in re ar Exam 1e 0. mg. 1 er s.

p Pro- Proofproduet of C3 pane pylene 00 Oxygen gas hydrocarbons consumed increased as the concentration of propylene oxide increased as the concentration of propylene was increased. It can also be seen from Examples 24-25 and 26-27 that increasing the oxygen decreases the efliciency.

pressure of 45 p.s.i.g., at a contact time of 12 seconds and with a feed composed of:

62 percent of propane by volume 20 percent of propylene by volume From this graph it can be seen in this process that operating at temperatures of around 525 C. will not only give a good yield of propylene oxide but will also give this propylene oxide in a process which produces less acetaldehyde than had heretofore been possible.

The same effect on acetaldehyde production is observed when the pressure is varied Within the critical limits of this invention. At the lower range of pressure the efficiency of the oxidation to (acetaldehyde is lower than at the upper range. "FIG. 13 depicts a graph of the etficiency of the oxidation reaction to acetaldehyde in relationship to the pressure. From this graph it can be seen why the preferred range of pressure is from 30-75 p.-s.1.g.

The following examples will illustrate the novel process of this invention.

Examples 1-18 These experiments were conducted in the same manner as Examples 1-18 except that the reactor had a volume of 500 milliliters. Operating conditions and results are also shown in Table IV.

Examples 26-27 These experiments were also conducted in the same manner as the preceding experiments except that the reactor had a volume of milliliters. Operating conditions and results are shown in Table IV.

It was HAC Ester r to the sampling ditions.

- A total of four thermo- M 12 ion zone, thermocouples were placed Example 36 CaHgO cr'raono HCHO CH4 les were used. The temperature was measured at Moi/ hr.

TABLE IV Mg./litcr of product gas Operating conditions 00 0.1160 omorro Horio co 00,

lso conducted in the same Examples 28-31 Examples 32-33 riments were a rirnents were conducted in the same manner -18 except that the partial pressure of the These expe These expe as Examples 1 manner as the preceding examples except that the reactor had a volume of 200 milliliters.

and results are also shown in Table IV.

reactants was held constant. Operating conditions and results are shown in Table IV. a

0 0 000000 0000000000006 w wwmmwmm1wm1w111111 111111111 B f l 6 26 29.222222880888824 .m 0 eoomfioonnooom 1 1111 111 1 1 t n 5 0 000000000000000 59080000000003 r a e C P l 8 8 28 28 9-84229-8883970822029-2281 M e 2 555-05555555555-0555555555555500050 444 44444444 4444444444444444470 4 P Lu-nv 77799922288820468M222229955555588 nm 00000011111111131 .11 .11 O l e r .1111100 .0000 .1 00 00 t 1) 30008412500021.25320789005512000000 00 0 0 0000999004557000000 homero mmsmmsess55425544445555? x 7 dnzaxiavaaorram 4 56 9 E Lifim hdnmlomflhmnm hflll1m122222222223333( Propane, pro- This gh transfer line The centrifugal During this t the total feed rate of gases to the reactor (A) Temperature measured by thermocouples extending (E) Total carbony1 compounds, calculated as formaldehyde into reaction zone. a in reactor fluent (0 C gglglteancegnrntsge hlgh boiling products separated by a watera water cgiggd tgltliarlhsieglh boiling products were separated by taighezi lrstillieacliltiiii cgn pounds, fialctulated as acetic acid conr 01 1D (C) Acetaldehyde contained in reactor efiiuent (0 C., 760 cooled condenser. S separated by a water mm. Hg), after high bOlllDg products were separated by a '(G). In this experiment 63.6 volume percent nitrogen was cooled condenser. added 'in the feed. (D) Formaldehyde contained in the high boiling products separatedby a water-cooled condenser. I

I Example 34 propylene oxide and other valuable oxygenated products.

. i ANALYSIS OF PRODUCTS IN THE VENTURLTYPE The reactoremployed had a volume of 0.00123 cubic foot REACTION VESSEL and was similar to thatshown in FIG. 3. py ene, and carbon monoxide Were measured by suitable order to detelmlne m6 d of mung m the back 55 flow meters into a common feed line to result in a mixfi s a 56158 of experiments were performed ture composed of 62 vol percent propane 20 vol perc nt an e ro ucts int c reaction zone were removed from U the react or at certain points along the reaction zone. pr-opylene and 10 percent mon9mde' In order to test the homo eneit OE the acfion mixture was fed through a 1 x 36-inch electrically heated g y l m 1 mm .tubular type heatexchanger whichpreheated the gas to ture under reaction conditions, propylene, propane, and 1 I o i def conditions Suitable for the approximate y 475 C. The preheated gas from the presamples of the matgri a1 heater was then fed tangentially throu 1 "2 int th i t were taken at four different points in the reactor, two of t mlxtng cone- 2 i 2%.. mac Where 11m oints were oneither side of the entrance nozzle, ea 6 Oxygen a approximate-y and ammmtmg thesalocations to 8 percent of the total reaction mixture was added to this rotating mass of gases through line 23 whmhwas the Contents in the reactor All Samples axially located at the base of the cone.

Sis Within experiment error thereby force caused intimate and instantaneous mixing of the demonstrating conclusively that this type reactor provides fi the l l i g i mlxtum Passed 3 complete mixing ofthe reactants and products. a Spmmng mo m e Sewn arger) cone- W are 1 1 .70 the temperature and the composltion of the mixture was Example 35 sufiicient to initiate and sustain the reaction. DETERMINATION OF 'lElMPERA'ZI?Ult IN THE experimen was 19.5 cubic feet per hour which corresponds to 21 ENE RI-TYPE RDA TI D EL l V U C ON V 4 residence time or 0.36 second at 45 p.s.1.g. pressure and 111 Order to Illustrate homogenelty of temperature at a temperature of 500 C. The reaction temperature (B) Propylene oxide contains wateroxygen were added un duction of propylene oxide.

these p and two were at the backof the reactor;

were considered optimum for determining the degree of homogeneity of had identical analy proportion to the heat produced by the reaction.

system pressure at 4-5 p.s.i.g.

was controlled by adjusting the temperature of the preheated gases, and by by-passing to 15 percent of the I was cooled to approximately 40 C. as it passed through a series of water-cooled condensers into a trap which removed relatively the high-boiling products which included formaldehyde in the form of a hydrate. The remaining gases containing most of the desired products. were measur ed by a gas meter and analyzed by chemical means and by a vapor-phase chromatograph. These analyses indicated a yield of propylene oxide amounting to 38 milligrams per liter of reactor etlluent which was obtained at an efliciency of 32 pounds per 100 pounds of propane consumed. .The yield of acetaldehyde was 30 milligrams per liter obtained at an etficiency of 17 pounds per 100 pounds of propane consumed.

Temperature measurements, made from thermocouples extending into various points within the reaction zone of the reactor, show variations of no more than 5 to 7 C. This indicates the reaction to be homogeneous in nature and that back-mixing was accomplished within the reactor. The back-mixing results from the entrance of the gas mixture into the space enclosed in the diverging walls of the cone. A low-pressure zone is thereby created near the apex of the cone and it causes the gas mixture near the base to return toward the apex via a current running axially at the center of the chamber. Thus, the reverse flow of gases causes the entire mass of gases to break up into an infinite number of small eddies and both intense mixing and back-mixing are achieved in this cone.

Example 37 VENTURLTYPE REACTOR The well-insulated reactor, corresponding to that shown a in FIGURE 2 had a volume of 3,540 cc. and the initial reaction temperature was obtained by preheating the inlet gas to such a temperature (approx. 400 C.) that the oxidation reaction was initiated. No external heating of the reactorwas necessary. Once operatingconditions had been attained, the preheater temperature was lowered in In recycle operation, the product gases, after cooling and scrubbing to separate products and by-products was mixed with fresh propane and propylene, and the enriched which has been shown in FIG. 14 (not drawn to scale) hydrocarbon feed, together with oxygen, was preheated to 320 C. before introduction into the reactor.

The reaction was carried out at 525 under substantially adiabatic conditions and essentially isothermal conditions, by maintaining a balance of volume of inlet gas (at 320 C.) such that the heat capacity of the cooler gas balanced the heat evolved in the exothermic reaction. To maintain these conditions, the inlet gas was mixed in the 'venturi mixer with approximately eight times its volume of reacting gases (at approximately 525 C.) in about one-tenth of the total residence time in the reactor. The reaction temperature was maintained by these means at 525i-12" C.

The product gas mixture was cooled quickly to room temperature; the condensable by-products were separated in a trap, at room temperature, and the desired reaction products were removed in a water scrubber. The scrubbed olf-gas was partially blown oli to maintain the The product gas contained, in addition to unreacted feed gas components, 1.39 volume percent total epoxide, 36.0 mg./l.; and 0.81 volume percent total aldehyde, 15.9 mg ./l. as determined by conventional analytical procedures. This corresponds to the following elficiencies:

26.8 lb. propylene oxide; 100 lb. of C hydrocarbon converted I 11.8 lb./aldehyde/ 100 lb. of C hydrocarbon converted at a 13.6% conversion of hydrocarbon per pass.

The feed gas composition, after make-up was:

The cycle gas volume, as determined after the traps to remove condensables and before the water scrubber, was 1,084 s.c.f.h. (at 21 C.), and the pressure drop through the nozzle of the venturi was 18 p.s.i.g.

Example 38 A horizontally disposed mechanically stirred reactor was constructed from a 3-inch section of 4-inch diameter,

N0. 316 stainless steel pipe 34 in which was axially disposed a 2-inch diameter pipe 33 enclosing a 2 inch diameter radial blower 35 operated at 3450 rpm. The reactor 34 had a volume of 390 cc. Feed gases entered the reactor through an annular preheater 31 designed to give a high preheating rate; the preheater was heated by means ofelectrical resistance winding 32. The feed gas was introduced-into the reactor Where the relatively large volfume of gas in the reactor under circulating conditions resulted in extremely rapid mixing so that the composition of the gas mixture at the reactor inlet was very nearly that of the product gas stream. Thus, there existed a steadystate gas composition in the reactor.

The reaction zone was maintained at 475 i3 by control of the exothermic reaction through regulating the temperature .of the inlet gas. The reactor pressure was '25 p.s.i.g., thecontact time was 0.84 second, and thecirculation rate of gas in the reactor Was 420 cubic ft./hr. The crude reaction product was cooled 36 before being passed 37 to traps to remove condensable by-products; then the gases were released by a back pressure controller, measured, analyzed, and then vented. The gaseous product, after passing through the traps, showed the following composition, by conventional methods of analysisz Propylene oxide, 0.68 volume percent of 15.0 mg./l. Aldehyde, 0.42 volume percent or 8.3 mg./l.; at a C hydrocarbon conversion of 2.5% per pass Feed gas composition: 1 1

Volume percent Example 39 Ethane was oxidized during single-pass operation in a double cone type of back-mixing reactor to produce ethylene oxide and othervaluableoxygenated products. The reactor employed had a volume of 0.00123 cubic foot and Ethane, ethylene,

meters into a common feed line to result in a mixture composed of 62 vol. percent ethylene and 10 vol. percent carbon monoxide. This mixture was fed .through a 1 x 36-inch electrically heated tubular type heat exchanger which preheated the gas to approximately 475 C. The preheated gas from the preheater was then fed tangentially through transfer line 21 into the mixing cone 22 of the reactor, where unheated oxygenat approximate 13/ 30 C. and amounting to 8 percent of'the total reaction mixture was added to this rotating mass of gases through line 23 which was axially located at the base of the cone. The centrifugal force caused intimate and instantaneous mixing of the gases, and the resulting homogeneous mixture passed With a spinning motion into the second (larger) cone 24 where the temperature and the composition of the mixture was suflicient to' initiate and sustain the reaction. During this experiment the total feed rate of gases to the reactor was such as as to provide a residencetime of 0.5 second at 45 p.s.i.g. pressure and at a temperature of 500 C. The reaction temperature was controlled by adjusting the temperature of the preheated gases, and by by-passing 10 to percent of the gases from the feed'line before the preheater into the second cone of the reactor through line 25.

When the reaction had reached a steady state, the effluent (product gas) leaving the reactor through line 26 was cooled to approximately 40 C. as it passed through a series of water-cooled condensers into a trap which removed relatively the high-boiling products which included formaldehyde in the form of a hydrate. The remaining gases containing most of the desired products were measured by a gas meter and analyzed by chemical its consisting essentially of at least one of said hydrocarbons, from 4 to 14 percent by volume of oxygen based on the volume of said hydrocarbon, and an inert gas,

(11) adjusting the rate of said feed so as to maintain a contact time of from 0.07 to 1.5 seconds in the reaction zone,

(c) maintaining said reaction zone at a pressure of from to 150 psig and at essentially isothermal condition within the range of 425 to 575 C.,

- (d) maintaining'said reaction zone under such condition of intermixing of the reactants and the reaction products that the concentrations of said reactants and reaction products remain substantially constant throughout said reaction zone, and

(e) recovering the reaction products from said reaction zone.

3. A method for the oxidation of hydrocarbons selected from the class consisting of ethane, propane, and butane 20 to produce the corresponding olefin oxide which method comprises:

(a) introducing into a reaction zone a feed mixture consisting essentially of at least one of said hydrocarbons and from 4 to 14 percent by volume of oxy- 1. A method for the oxidation of hydrocarbons selected from the class consisting of ethane, propane, and butane, to produce the corresponding olefinoxide, which method comprises: 7 v I (a) introducing into a reaction zone a feed mixture consisting essentially of at least one of said hydrocarbons and from 4 to 14 percent by volume of oxygen based on the volume of said hydrocarbon,

(b) adjusting the rate of said feed so as to maintain a contact time of from 0.07 to 1.5 second in the reaction zone,

(c) maintaining said reaction zone at a pressureiof from 20 to 150 p.s.i.g. and at essentially isothermal condition within the range of 425 to 575 C.,

(d) maintaining said reaction zone under such condi-.

tion of intermixing of the reactants and the reaction products that the concentrations of said reacttants and reaction products remain substantially constant throughout said reaction zone, and (e) recovering the reaction products from said reaction Z0116. i 2. A method for the oxidation of hydrocarbons selected from the class consisting of ethane, propane, and butane, to produce the corresponding olefin oxide, which method comprises: Y

(a) introducing into a reaction zone a feed mixture means and by a vapor-phase chromatograph. These gen based on the volume of said hydrocarbon, analyses indicated a yield of ethylene oxide amounting to (b) ad usting the rate of said feed so as to maintain 9.5 milligrams per liter of reactor effluent which was 0b a contact time of from 0.07 to 1.5 seconds in the tained at an efficiency of 20.9 pounds per 100 pounds of reactlonzone, hydrocarbonconsumed. The yield of aldehyde was 7.4 (c) maintaining said reaction zone at a pressure of milligrams per liter obtained at an efi'i ieney' of 16,5 from 20 to 150 p.s.1.g. and at essentially isothermal pounds per 100 pounds of'hydrocarbon consumed. COHditlOH Withm the range of 425 to 1 (d) maintaining saidreaction zone under condition Exampl f of intermixing of the reactants and the reaction prod- These expernnents were carried out 1n a similar manucts that the concentrations of said reactants and rener as Example 34. Operatmg conditions and results action products remain substantially constant a are shown in Table V. throughout said reaction zone,

TABLE v Volumef peircent in Product cone. Products in 1bs./100lbs. of C2 hydrocarbons consumed 0G Ex. C. (tlpnt. P.s.i.g.

ime

' 02m CZHA 02 CO CZHdO, Ald., HCHO, c0 00, Cal-I40 Ald. HOHO CH4 o; 01 HAC Ester mg./l. mg./l. moL/hr.

40.. 450 0.5 02 20 s 10 8.4 8.2 1.41 80.4 18.9 13.9 13.5 33.3 5.5 3.9 2.2 3.2 5.5 41 450 0.5 45 72 20 s 0 8.8 8.8 1.70. 01.0 12.6 15.3 14.0 41.0 2.4 5.2 2.4 4.2 9.7 42 500 0.5 45 '72 20 s 0 11.9 7.5 1.19 47.2 28.4 21.6 13.4 31.1 7.7 8.0 3.1 7.8 7.3 43-- 525 0.5 45 72. 20 s 0 15.3 7.2 1.00 46.2 30.8 24.0 11.3 21.4 13.0 6.6 3.9 6.5 0.0

What is claimed: (e) continuously recycling any unreacted hydrocarbons back into said'feed so that said feed mixture enters the reaction zone at a temperature which is sufficiently lower than the reaction temperature so as to maintain substantially adiabatic condition in thereaction zone, and i V (f) recovering the reaction products from said reaction zone. V 4. The method of claim 3 wherein the contact time is from 0.10 to 0.6 second, the temperature of the reaction zone is from 450 to 550 C., the pressure is from 30 to p.s.i.g. and the amount of oxygen in the feed is from 6 to 8 percent by volume.

5. A method for the oxidation of hydrocarbons selected from the class consisting of ethane, propane, and butane to produce th'e'corresponding olefin oxide which method comprises:

(a) introducing into a reaction zone a feed mixture consisting essentially of at least one of said hydrocarbons, from 4 to 14 percent by volume of oxygen based on the volume of said hydrocarbon, and an inert gas,

' (b) adjusting the rate of said feed so as to maintain a contact time of from 0.07 to 1.5 seconds in the reaction zone,

(c) maintaining said reaction zone at a pressure of from 20 to 150 p.s.i.g. and at essentially isothermal condition within the range of 425 to 575 C.,

(d) maintaining said reaction zone under such'condition of intermixing of the reactants and the reaction products that the concentrations of said reactants and reaction products remain substantially constant throughout said reaction zone,

(e) continuously recycling any unreacted hydrocarbons back into said feed so thatsaid feed mixture enters the reaction zone at a temperature which is sufficiently lower than the reaction temperature so as to maintain substantially adiabatic condition in the reaction zone, and

(f) recovering the reaction products from said reaction zone.

6. A process for the oxidation of hydrocarbons selected from the class consisting of ethane, propane, and butane to the corresponding olefin oxide which process comprises:

(a) introducing into a reaction zone a feed mixture consisting essentially of at least one of said hydrocarbons, an olefin which has the same number of carbon atoms as the corresponding hydrocarbon and oxygen,

(b) adjusting the rate of said feed so as to maintain a contact time of from 0.07 to 1.5 seconds in the reaction zone,

(d) maintaining a ratio of said hydrocarbon to said olefin in the feed of from 1:2 to :1 and an oxygen concentration of from 4 to 14 mol percent of the entire feed,

(e) maintaining said reaction zone under such condition of intermixing of the reactants and the reaction products that the concentrations of said reactants and reaction products remain substantially constant throughout said reaction zone, and

(f) recovering the reaction products from said reaction zone.

'7. A process for the oxidation of hydrocarbons selected from the class consisting of ethane, propane, and butane to the corresponding olefin oxide which process comprises:

(c) maintaining :said reaction zone at a pressure of from 20 to 150 p.s.i.g. and at essentially isothermal condition Within the range of 425 to 575 C.,

(d) maintaining a ratio of said hydrocarbon to said olefin in the feed of from 1:2 to 10:1 and an oxygen concentrationof from 4 to 14 mol percent of the entire feed,

(f) recovering the reaction products from said reaction zone. 1

8. The process of claim 7 wherein the contact time is from 0.10 to 0.6 second, the temperature of the reaction zone is from 450 to 550 C., the mol ratio of hydrocarbon to olefin is from 3:1 to 4:1 and the amount of oxygen in the feed is from 6 to 8 percent.

9. A process for the oxidation of propane to propylene oxide which comprise-s:

(a) introducing into a consisting essentially of propane and from 4 to 14 percent by volume of oxygen based on the volume of said propane,

(b) adjusting the rate of said feed so as to maintain a contact time of from 0.07 to 2 seconds,

(c) maintaining said reaction zone at a pressure of from 20 to p.s.i.-g. and at essentially isothermal condition within the range of 425 to 575 C.,

(d) maintaining said reaction zone under such condi tion of inter-mixing of the reactants and the reaction products that the concentrations of said reactants and reaction products remain substantially constant throughout said reaction zone, and

(e) recovering the reaction products from said reaction zone.

10. Aprocess for the oxidation of propane to propylene oxide which comprises:

(a) introducing into a reaction zone a feed mixture consisting essentially of propane and from 4 to 14 percent by volume of oxygen based on the volume of propane,

(0) maintaining said reaction zone at a pressure of from 20 to 150 p.s.i.g. and at essentially isothermal condition within the range'of 425 to 575 C.,

(d) maintaining said reaction zone under such condition of intermixing of the reactants and the reaction products that the concentrations of said reactants and reaction products remain substantially constant throughout said reaction zone,

(e) continuously recycling any unreacted hydrocarbons back into said feed so that said feed mixture enters into the reaction zone at a temperature which is sufliciently lower than the reaction temperature so as to maintain substantially adiabatic condition in the reaction zone, and

(f) recovering the reaction products from said reaction zone.

11. The process of claim 10 wherein the contact time is from 0.10 to 0.6 second, the temperature of the reac tion zone is from 450 to 550 C., the pressure is from 30 to 75 psig and the amount of oxygen in the feed is from 6 to 8 percent by volume. r

12. A process for the oxidation of propane to propylene oxide which comprises:

(a) introducing into a reaction zone a feed mixture consisting essentially of propane, propylene, and oxygen,

(0) maintaining said reaction zone at a pressure of from 20 to 150 p.s.i.g. and at essentially isothermal condition within the range of 425 to 575 C.,

(d) maintaining a ratio of said hydrocarbon to said olefin of from 1:2 to 10:1 and an oxygen concentration of from 4 to 14 percent by volume of the entire feed,

(e) maintaining said reaction zone under such condition of intermixing of the reactants and the reaction products that the concentrations of said reactants and reaction products remain substantially constant throughout said reaction zone,

(f) continuously recycling any unreacted propane back into the feed so that said feed mixture enters the reaction zone at a temperature which is sufiiciently lower than the reaction temperature so as to maintain substantially adiabatic condition in the reaction zone, and

(g) recovering the reaction products from said reaction zone.

(References on following page) reaction zone a fed'mixture Gardner et a1. '5. Dec. 25, 1956 Steitz Jan. 5, 1960 V Lalng 'et a1 Apr. 25, 1961 OTHER REFERENCES Satterfield et a1.: Industrial and Engineering Chemistry, ol. 46, No. 5 (May 1954 pp. 1001-1007 (pp. 1001 Jones et 211.: Ind: & Eng. Chem, vol. 51, pp. 262, 263,

UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Patent No. 3,132,156 May 5, 1964 Russel C. Lemon et a1.

It is hereby certified that error appears in the above numbered patent requiring correction and that the said Letters Patent should read as corrected below.

Column 2, lines 16 to 22, the reactions should appear as shown below ins tead of as in the patent:

intermediates- 9 carbon oxides, water,

aldehydes, etc.

Signed and sealed this 29th day of September 1964.

(SEAL) Attest:

ERNEST W. SWIDER EDWARD J. BRENNER Attesting Officer Commissioner of Patents

Claims

1. A METHOD FOR THE OXIDATION OF HYDROCARBONS SELECTED FROM THE CLASS CONSISTING OF ETHANE, PROPANE, AND BUTANE, TO PRODUCE THE CORRESPONDING OLEFIN OXIDE, WHICH METHOD COMPRISES: (A) INTRODUCING INTO A REACTION ZONE A FEED MIXTURE CONSISTING ESSENTIALLY OF AT LEAST ONE OF SAID HYDROCARBONS AND FROM 4 TO 14 PERCENT BY VOLUME OF OXYGEN BASED ON THE VOLUME OF SAID HYDROCARBON, (B) ADJUSTING THE RATE OF SAID FEED SO AS TO MAINTAIN A CONTACT TIME OF FROM 0.07 TO 1.5 SECOND IN THE REACTION ZONE, (C) MAINTAINING SAID REACTION ZONE AT A PRESSURE OF FROM 20 TO 150 P.S.I.G. AND AT ESSENTIALLY ISOTHERMAL CONDITION WITHIN THE RANGE OF 425 TO 575*C., (D) MAINTAINING SAID REACTION ZONE UNDER SUCH CONDITION OF INTERMIXING OF THE REACTANTS AND THE REACTION PRODUCTS THAT THE CONCENTRATIONS OF SAID REACTTANTS AND REACTION PRODUCTS REMAIN SUBSTANTIALLY CONSTANT THROUGHOUT SAID REACTION ZONE, AND (E) RECOVERING THE REACTION PRODUCTS FROM SAID REACTION ZONE.